diff --git a/src/content/skills-zh/wechat-10w-article.md b/src/content/skills-zh/wechat-10w-article.md
new file mode 100644
index 0000000..eceddb5
--- /dev/null
+++ b/src/content/skills-zh/wechat-10w-article.md
@@ -0,0 +1,45 @@
+---
+name: gzh-10w-article-recommend
+title: 公众号热门文章 TOP10 推荐
+description: 根据用户输入关键词调用脚本，获取公众号热门文章数据，输出可读的 TOP10 文本榜单并生成 HTML 卡片预览。
+source: community
+author: yuanyi-github
+githubUrl: https://github.com/yuanyi-github/skills
+category: marketing
+tags:
+  - 公众号
+  - 热门文章
+  - 榜单推荐
+  - html预览
+roles:
+  - assistant
+featured: false
+popular: false
+isOfficial: false
+installCommand: |
+  git clone https://github.com/yuanyi-github/skills
+  cp -r skills/gzh-10w-article-recommend ~/.qoder/skills/
+date: 2026-04-27
+---
+
+## 使用场景
+
+- 用户希望按关键词查看公众号热门文章并获得可参考的 TOP10 榜单
+- 需要同时得到文本输出与 HTML 卡片化预览，便于快速浏览与分享
+- 需要按阅读数排序并携带点赞、评论、分享等互动指标
+
+## 示例
+
+```bash
+python scripts/fetch_explosive_articles.py \
+  --origin_word "AI编程" \
+  --spit_words '["AI","编程"]' \
+  --expansion_words '[]'
+```
+
+## 注意事项
+
+- 严格执行完整流程：接收输入、分词、调用脚本、输出文本榜单、生成并展示 HTML
+- 文本榜单与 HTML 预览必须同时提供，不可只输出其中一种
+- 榜单以脚本返回数据为准，通常展示阅读数靠前的前 10 条，不足则按实际数量展示
+- 互动与阅读数据可能存在入库延迟，非实时口径
diff --git a/src/content/skills/wechat-10w-article.md b/src/content/skills/wechat-10w-article.md
new file mode 100644
index 0000000..c1a9627
--- /dev/null
+++ b/src/content/skills/wechat-10w-article.md
@@ -0,0 +1,45 @@
+---
+name: gzh-10w-article-recommend
+title: Top 10 WeChat Popular Article Recommendations
+description: Calls a script based on user keywords to fetch popular WeChat article data, outputs a readable Top 10 text ranking, and generates an HTML card preview.
+source: community
+author: yuanyi-github
+githubUrl: https://github.com/yuanyi-github/skills
+category: marketing
+tags:
+  - WeChat Official Accounts
+  - Popular Articles
+  - Ranking Recommendations
+  - HTML Preview
+roles:
+  - assistant
+featured: false
+popular: false
+isOfficial: false
+installCommand: |
+  git clone https://github.com/yuanyi-github/skills
+  cp -r skills/gzh-10w-article-recommend ~/.qoder/skills/
+date: 2026-04-27
+---
+
+## Use Cases
+
+- Users want to view popular WeChat articles by keyword and get a referenceable Top 10 ranking.
+- Users need both text output and HTML card-based preview for quick browsing and sharing.
+- Users need ranking by read count with engagement metrics such as likes, comments, and shares.
+
+## Example
+
+```bash
+python scripts/fetch_explosive_articles.py \
+  --origin_word "AI编程" \
+  --spit_words '["AI","编程"]' \
+  --expansion_words '[]'
+```
+
+## Notes
+
+- Follow the complete workflow strictly: receive input, tokenize, call the script, output the text ranking, then generate and display HTML.
+- The text ranking and HTML preview must both be provided; do not provide only one.
+- The ranking is based on script-returned data, usually showing the top 10 by read count; if fewer are available, show the actual number.
+- Engagement and read metrics may have ingestion delay and are not real-time values.